Learning the Regularization Strength for Deep Fine-Tuning via a Data-Emphasized Variational Objective

Re-assign community